skills/xsavikx/okf-skills/okf-git/Gen Agent Trust Hub

okf-git

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill performs legitimate repository analysis and documentation tasks. It accesses local repository metadata and file structures to generate documentation.
  • [EXTERNAL_DOWNLOADS]: The installation instructions use the standard Go toolchain to fetch the connector from the author's repository.
  • [COMMAND_EXECUTION]: The skill requires manual execution of commands by the user to build and run the connector binary.
  • [DATA_EXFILTRATION]: File access is limited to the targeted Git repository and user-defined output paths. The skill respects .gitignore and .okfignore files to avoid processing restricted files.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Git repositories to generate summaries. This is treated as safe as it is the primary intended function and lacks autonomous execution of instructions found in processed data.
  • Ingestion points: Local file contents and commit history in main.go and relationships.go.
  • Boundary markers: No explicit delimiters found in the connector logic.
  • Capability inventory: File system write access for bundle creation and metadata sync.
  • Sanitization: Relies on internal library functions for Markdown generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 07:05 PM
Security Audit — agent-trust-hub — okf-git